besides purkinje cells and granule neurons: an appraisal ... · located in the granule cell layer...

Histochem Cell Biol (2008) 130:601–615
DOI 10.1007/s00418-008-0483-y
REVIEW

Besides Purkinje cells and granule neurons: an appraisal of the cell biology of the interneurons of the cerebellar cortex

Karl Schilling · John Oberdick · Ferdinando Rossi · Stephan L. Baader

Accepted: 15 July 2008 / Published online: 2 August 2008© The Author(s) 2008

Abstract Ever since the groundbreaking work of Ramony Cajal, the cerebellar cortex has been recognized as one ofthe most regularly structured and wired parts of the brainformed by a rather limited set of distinct cells. Its ratherprotracted course of development, which persists well intopostnatal life, the availability of multiple natural mutants,and, more recently, the availability of distinct moleculargenetic tools to identify and manipulate discrete cell typeshave suggested the cerebellar cortex as an excellent modelto understand the formation and working of the central ner-vous system. However, the formulation of a unifying modelof cerebellar function has so far proven to be a most cantan-kerous problem, not least because our understanding of theinternal cerebellar cortical circuitry is clearly spotty. Recentresearch has highlighted the fact that cerebellar corticalinterneurons are a quite more diverse and heterogeneousclass of cells than generally appreciated, and have providednovel insights into the mechanisms that underpin the devel-opment and histogenetic integration of these cells. Here, weprovide a short overview of cerebellar cortical interneuron

diversity, and we summarize some recent results that arehoped to provide a primer on current understanding of cere-bellar biology.

Keywords Interneuron · Basket cell · Stellate cell · Lugaro cell · Unipolar brush cell · Candelabrum cell · Development

Introduction

“The cerebellar cortex is built from four main types of neu-rons: granule cells, Purkinje cells and two types of inhibi-tory interneurons, Golgi cells and the stellate/basket cells”(Voogd and Glickstein 1998): To this day, this statement,or any conceivable permutation or paraphrase of it, is a cor-nerstone of most descriptions introducing the basic anat-omy of the cerebellum. Yet it is well-known that besidesthe four neuronal phenotypes just mentioned, the cerebellarcortex contains several other types of neurons, among themcandelabrum cells, Lugaro cells, and unipolar brush cells.The fact that these latter cells are usually not mentioned inthe standard introductory phrase to cerebellar histology (butsee Rong et al. 2004 for a rare exception), or even standardtextbooks, reXects nothing less than the fact that our knowl-edge of the existence of these cells, not to mention theirfunction, is of relatively recent vintage, and often ratherfragmentary.

Even among the better known constituents of cerebellarcortical circuitry, basket/stellate and Golgi cells may be setapart from Purkinje and granule cells by the fact that theirdevelopmental history, in particular, is only beginning toemerge. The availability of a number of natural and engi-neered mutants aVecting primarily and/or directly Purkinjeand/or granule cells have been known for quite some time

K. Schilling (&) · S. L. BaaderAnatomisches Institut, Anatomie und Zellbiologie, Rheinische Friedrich-Wilhelms-Universität, Nussalle 10, 53115 Bonn, Germanye-mail: [email protected]

J. OberdickDepartment of Neuroscience, Center for Molecular Neurobiology and Department of Neuroscience, The Ohio State University, Columbus, OH, USA

F. RossiDepartment of Neuroscience, “Rita Levi Montalcini Centre for Brain Repair”, National Institute of Neuroscience, University of Turin, Torino, Italy

123

602 Histochem Cell Biol (2008) 130:601–615

(for reviews and further references, see, e.g., Caviness andRakic 1976; Goldowitz and Hamre 1998; Oberdick et al.1998; Sotelo 2004) and provided an inroad to unravel basicaspects of the diVerentiation, function and systemic signiW-cance of these cells. In contrast, genetic and molecularmeans to identify, characterize, or manipulate basket, stel-late and Golgi cells, candelabrum neurons, Lugaro cells orunipolar brush cells have been largely elusive. It is not sur-prising, then, that these cells are typically not consideredeven in recent integrative views of cerebellar function (e.g.,Apps and Garwicz 2005; Porrill et al. 2004; Boyden et al.2004); at best, the more prominent Lugaro cells and unipo-lar brush cells are mentioned, and rather in passing (e.g., Ito2008; Millen and Gleeson 2008).

Yet recently, a combination of classical morphologicaland molecular approaches has provided novel means tounravel the biology of these cells. Together, they constitutean important step in our understanding of the cellular make-up of the cerebellum and towards a systems-level under-standing of how the functions of the cerebellar cortex aremechanistically implemented. Here, we try to summarizesome of the more recent advances in our knowledge of thedevelopmental history (Fig. 1), molecular makeup, and

synaptic wiring (Fig. 2) of these less well-known constitu-ents of the cerebellar cortex.

Unipolar brush cells

Of all cerebellar cortical interneurons, unipolar brush cellslocated in the granule cell layer stand out as the only excit-atory neurons. While they have been described only ratherrecently, their cell and developmental biology is by nowrather well deWned. In 1977, Altman and Bayer Wrstdescribed a cerebellar neuron, located in the granule celllayer and preferentially found in the nodulus, whichdiVered from granule or Golgi cells by its pale nucleus andits date of generation (Altman and Bayer 1977). By allknown criteria, the cells then described seem to correspondto a subset of what is now known as unipolar brush cells.This descriptive name was coined by Mugnaini and associ-ates, who also provided the Wrst detailed description ofthese cells (Harris et al. 1993; Mugnaini and Floris 1994).

By combining the observations of Altman and Beyer(1977); cf also Fig. 22–23 and 22–24 in Altman and Bayer1997) and Sekerkova et al. (2004), one may estimate that,within the vermis some 42% of all UBCs are localized inthe Xocculus (vermis of lobule X), and some 24% in theuvula (vermal part of lobule IX); the rest is distributedthroughout other parts of the cerebellum. Further, thesestudies suggest that within the nodulus, the ratio of UBCsto Pjs is about 3:1, and in the declive (vermis of lobus VI)0.3:1.

Unipolar brush cells are rather small neurons with a sin-gle, quite thick but stubby dendrite which terminates in abrush-like spray of dendrioles. Consequently, they appearas electronically compact neurons. UBCs are innervated bya single mossy Wber (MF), providing vestibular aVerents,which makes contact with the entire dendritic brush andforms an unusually large synapse comprising multiple pre-synaptic release sites apposed to continuous regions ofpostsynaptic densities (Mugnaini et al. 1994; Rossi et al.1995). This synapse is endowed with ionotropic, but lacksmetabotropic glutamate receptors (Jaarsma et al. 1995).The latter, however (mGluR1, mGluR1alpha and mGluR2/3 immunoreactivity) are found in extrasynaptic membranesand densely localized to non-synaptic appendages of theUBC dendrioles (Mugnaini et al. 1997 and references citedtherein). (Vestibulocerebellar) UBCs are also innervated byGolgi cells, which co-release GABA and glycine at thesesynapses. The UBC response, however, was found to beeither mixed GABAergic and glycinergic, or purely glycin-ergic (Dugue et al. 2005). It is currently not known whetherdiVerences in GABAA-receptor mediated responsiveness ofUBCs reXect functional heterogeneity, or are due to diVer-ential maturation/diVerentiation of UBCs at postnatal days

Fig. 1 A schematic view of histogenetic events in the early postnatal(in the mouse, say, up to postnatal day 15) cerebellar cortex. Granulecell precursors proliferate in the proliferative part of the external gran-ule cell layer (egl). The actual cell cycle duration varies with age, butis in the range from 15 to 29 h. After their last mitosis, granule cellsremain in the inner part of the egl, where they start to elaborate neu-rites, for about 28 h, before they rapidly (within some 4 h) migrate totheir Wnal position in the inner granule cell layer and start to elaboratedendrites. Granule cell development is shown in red/yellow. In con-trast, precursors of inhibitory interneurons (labeled greenish) reach thecerebellar cortex through the nascent white matter. They seem to freelytraverse the nascent (internal) granule cell layer, but do not penetrateinto the external granule cell layer. In contrast to our detailed knowl-edge of the kinetics and dynamics of granule cell formation and migra-tion, hardly and details of the time course of the diVerentiation ofinhibitory interneurons are known. For further details and additionalreferences, see Fujita (1967); Sotelo (2004); Gliem et al. (2006); We-isheit et al. (2006)

123

Histochem Cell Biol (2008) 130:601–615 603

17–21, when these recordings were acquired (from ratmaterial). Finally, expression of PX2 purinergic receptorshas also been observed in developing UBCs (Xiang andBurnstock 2005). While a dedicated source of ATP-ligandis currently unknown, this may form a basis of glia medi-ated modulation of these cells, somewhat analogous to thesituation described for Purkinje (Brockhaus et al. 2004) orLugaro cells (cf above; Saitow et al. 2005).

From the perikaryon, UBCs emanate a single axon, whichtakes a tortuous course within the granule cell layer andoccasionally may course for a short distance through thewhite matter. Eventually, it branches repeatedly in the gran-ule cell layer and forms a plexus displaying rosette-like exc-rescences that form the central component of glomeruli. Inthe latter, UBC axons are surrounded by granule cell den-drites and putative Golgi cell dendrites. Structurally, thesynapses in these rosettes appear asymmetric (Berthie andAxelrad 1994; Rossi et al. 1995). Unipolar brush cells areglutamatergic (Nunzi et al. 2001), and given their wiring,they may contribute to a patterned spread of vestibular aVer-ent excitation within the granule cell layer that has been con-sidered to be equivalent to a feed-forward excitation (Dinoet al. 2000). Their primary function seems to be to coordi-nate the synchronized activity of sets of granule cells, whichin turn would regulate activity in spatially restricted subsetsof Purkinje cells (cf Nunzi et al. 2001 for further details).

Among the molecularly distinguishing features of unipo-lar brush cells are the relative abundance of (partially non-phosphorylated), high molecular weight neuroWlament(NF-H), which presumably also forms the epitope recog-nized by monoclonal antibody Rat-302 (Harris et al. 1993).Attempts to deWne UBCs molecularly have also shown thatthere exist several partly overlapping subsets of UBCs, asdeWned by their diVerential expression of a panel of mark-ers, including calretinin, the metabotropic glutamate recep-tor subunit 1� (mGluR1�), ionotropic glutamate receptorsubunits 2 and/or 3 (GluR2, Glur2/3), glutamate transport-ers Vglut1 and Vglut2, secretogranin and chromogranin A(Jaarsma et al. 1995; Floris et al. 1994; Nunzi et al. 2002;Nunzi et al. 2003; cf also Vig et al. 2005b and referencestherein). Yet none of these markers is speciWc; moreover,there seems to exist pronounced species diVerences withrespect to the expanse of UBC-subpopulations expressingthese markers or combinations thereof. The physiologicconsequences of this diversity are not understood.

Distinct subsets of UBCs may also be deWned based onwhen these cells go through their last mitosis (their “birth-date”). In the rat, this occurs from embryonic day 15 (i.e.,starting at about the time the last Purkinje cells are gener-ated) and into the early postnatal period, up to about post-natal day 2. Those that will eventually co-express calretininand the ionotropic glutamate receptor subunit 2 (GluR2) go

Fig. 2 A schematic and simpliWed view of the neurons of the cerebel-lar cortex and their wiring. For simplicity, neurochemically deWnedsubsets of Golgi cells are shown as one single cell type (Go). InhibitoryGolgi (G), basket (B), stellate (S) are shown in green; inhibitory Lug-aro cells (L) are shown in light (classical Lugaro cells) or dark blue(globular tye). Excitatory granule cells are labeled red, and excitatoryunipolar brush cells (U) are marked orange. Climbing (CF) and mossy

Wbers (Mo) are also excitatory. The transmitter and wiring of candela-brum cells (Ca) is not yet known, though their axons project into themolecular layer. P Purkinje cells, ser serotoninergic aVerents while thisscheme shows how individual cell types are interconnected, it does notconvey details of the three-dimensional details of these interconnec-tions, which is brought about by the often virtually two-dimensionalexpanse of the dendrites and/or axons of individual cell types

123


through their last mitosis, in between E15 and E21, with apronounced peak at E17/18. Calretinin-negative cells,which seem to be identical to the pale cells of Altman andBayer (1977), do so from E17 to E22, and into the Wrst twopostnatal days, with a broad peak from E19 to E21 (Sekerk-ova et al. 2004; Altman and Bayer 1977). Finally, as shownin mice, UBC subsets deWned by diVerential staining forcalretinin and mGluR1� receive diVerent sets of mossyWber aVerents, and this may indeed direct their actual neu-rochemical diVerentiation (Nunzi et al. 2002).

The recent observation that UBCs and their precursorsexpress the transcription factor Tbr2 allowed for the Wrsttime to locate and follow these cells in the cerebellar anlage(Englund et al. 2006). These studies showed that precursorsof UBCs can be localized to the rhombic lip in the early (E13.5 in the mouse) cerebellar anlage, from where theytranslocate into the cerebellar granule cell layer (cortex) viathe nascent cerebellar white matter. Thus, these studies vin-dicate the observations of (Takacs et al. 2000) who hadreported the presence of UBCs in the white matter of imma-ture cerebellum. One caveat to heed is that we do not knowwhether all UBC-precursors are indeed Tbr2-positive.

While the work of Englund et al. (2006) clearly docu-ments that the numerical expansion of the Tbr2-positivepopulation is critically dependent on Math1 expression, it isless obvious whether this reXects a cell-intrinsic depen-dency, or rather an early regulatory interaction between theMath1-positive granule cell lineage and that of UBCs.Thus, while the authors report weak expression of ß-galac-tosidase from the Math1 locus in Tbr2 positive cells, itremains open whether this is indeed indicative of expres-sion of cognate Math1 in these cells, which would also besuggestive of a molecular relationship between these lin-eages. Indeed, data reported by Zervas et al. (2004) maysuggest that precursors of UBCs may be distinguished fromthe granule cell lineage as early as E10.5 based on diVeren-tial expression of wnt-1. These authors describe, in mice inwhich wnt-1 expressing cells were genetically marked atE10.5, descendants of these cells that, within the cerebel-lum, localized exclusively into the granule cell layer of lob-ules IX and X. While the data shown do not really allowthese cells to be identiWed, their distribution is highly sug-gestive of unipolar brush cells, and clearly raises the issueas to the existence of an independent lineage as early asE10.5 in the mouse. This interpretation would also be con-sistent with the conclusion reached by Englund et al. (2006)that UBCs originate from the rhombic lip, as wnt-1 expres-sion within the cerebellar anlage is restricted to this struc-ture at E 10.5 (i.e. at the time the putative UBCs describedby Zervas et al. were genetically marked; Li et al. 2002).

While it is clear that UBCs migrate through thenascent white matter, the mechanisms that direct theirmigration and assure their preferential localization in the

vestibulocerebellum are not known. While UBC position-ing is disturbed in reeler mice (Ilijic et al. 2005), it is notdecided whether this is a direct eVect or rather a reXectionof the general disorganization of the cerebellar cortex inthese mutants (Ilijic et al. 2005; Englund et al. 2006).

Following their last mitosis, UBCs acquire their charac-teristic and name-giving morphology over a rather pro-tracted period, during which they also become synapticallyinvested. This has been described in detail for the rat byMorin et al. (2001) and for cat by Takacs et al. (2000). Isthe normal diVerentiation of UBCs related to their synapticintegration? In reeler mice, a striking segregation betweenthe localization of (calretinin-positive) UBCs, which arefound displaced in the ventroposterior part of the cerebel-lum, and the terminal Welds of secondary vestibulocerebel-lar aVerents is observed (Vig et al. 2005a). Yet theseWndings are hard to interpret with respect to the interdepen-dence of UBC positioning and aVerent innervation, for tworeasons: as pointed out by Vig et al. (2005a), the UBCsthey identiWed could still be innervated by primary vestibu-locerebellar aVerents; and, calretinin-negative UBCs(which form the majority of all UBCs in the rat; mouse datanot available) were not investigated. Indeed, Ilijic et al.(2005) have reported that UBCs in reeler mice form synap-tic junctions with complex axon terminals, possibly repre-senting mossy Wbers and UBC axons, just like UBCs inwildtype animals. If there was any diVerence betweenectopic UBCs in reeler and their normally situated counter-parts in non-mutant animals, these seemed to be limited to asomewhat looser brush-structure of their terminal dendrites(Ilijic et al. 2005). In primary cerebellar cultures, whichlack extracerebellar mossy Wbers, UBCs develop and maybe recognized based on their expression of calretinin. Whilesome cultured UBCs develop the name-giving morphologyof these cells, that of most cells is suYciently variable toargue in favour of a strong inXuence of orderly local cuesfor the proper morphogenesis of these cells (Anelli et al.2000; Anelli and Mugnaini 2001). We do not knowwhether and how aVerent innervation impinges on the neu-rochemical diVerentiation and diversiWcation of UBCsmentioned above.

Candelabrum cells

The candelabrum cell was Wrst described in 1994 by Laineand Axelrad (1994) in the rat, and is the most recentlydelineated distinct neuronal phenotype of the cerebellarcortex. Candelabrum cell perikarya are located within theganglionic (Purkinje) cell layer, and they are usually elon-gated along the vertical axis. Typically, these cells have oneor two long, rarely branched dendrites, which ascendalmost vertically into the molecular layer, and several (3–5)

123


short dendrites which project into the granule cell layer,where they run preferentially in the horizontal plane(Fig. 2). Both types of dendrites are covered with spines.The candelabrum cell axon projects into the molecularlayer, where it runs horizontally and emits multiple beadedbranches, which ascend vertically through most of themolecular layer, where they are preferentially arranged inparasagittal planes. This axonal plexus may either occupy aterritory that essentially overlaps with the area occupied bythe same cell’s ascending dendrites, or be displaced later-ally some variable distance. Candelabrum cells are distrib-uted throughout all parts of the cerebellar cortex atapparently roughly equal density.

While the dendritic structure of candelabrum cells suggestthat parallel Wbers, ascending granule cell axons, but alsobasket/stellate cells and climbing Wbers may provide aVerentinput to them, their aVerents have so far not been identiWed.The same holds for its target(s), although the axonal struc-ture makes Purkinje cell dendrites (and those of basket/stel-late cells) likely targets. While their location and details oftheir dendritic and axonal morphology certainly distin-guishes candelabrum cells from basket and stellate cells,their presumed wiring also suggests that they may be part ofan extended basket/stellate/candelabrum cell classiWcation.This might suggest a transmitter phenotype and also hint attheir potential developmental history. Indeed, candelabrumcells in the monkey (Macaca) cerebellum have recently beenshown to be immunoreactive for glycine, GABA, and GAD(the GABA-synthesizing enzyme glutamate decarboxylase;Crook et al. 2006), an observation in accord with the previ-ous tentative identiWcation, in rats, of cells doubly reactivefor GAD67 and the glycine transporter, Glyt2, as possiblycomprising candelabrum cells (Tanaka and Ezure 2004). Inhuman cerebellum, a cell-type which based on its locationand the shape of its perikaryon may be classiWed as candela-brum cells were found to be GAD65/67 (the antibody useddid not allow to distinguish isoforms; Flace et al. 2004).Thus, current evidence concurs to indicate that candelabrumcells use GABA and glycine as transmitters.

The work of Crook et al. (2006) also indicates that can-delabrum are rather sparsely contacted by GAD-immuno-positive presynaptic elements. However, we are essentiallyignorant as to the synaptic investment of candelabrum cells,their aVerents, their receptor endowment, let alone theirelectrophysiological properties.

To date, we know of no reliable and speciWc molecularmarker for candelabrum cells. We have no numerical esti-mates as to their prevalence. Their developmental history iscompletely obscure. Thus, we do not know whether Cande-labrum cells, or their precursors, are positive for ROR�and/or Pax-2; clariWcation of these points, should help toclarify whether and how Candelabrum cells relate to Golgi,basket and stellate cells (see also below).

Large inhibitory interneurons of the granule cell layer

Often, large inhibitory interneurons in the granule cell layerhave been equated with Golgi neurons; yet it was CamilloGolgi (1903) himself who Wrst drew attention to a class ofcells distinct from those that later were to be called by hisname, and which are now known as Lugaro cells. Thisname acknowledges the Wrst detailed description of thesecells by Ernesto Lugaro (1894). Moreover, and signiW-cantly, recent research has unveiled an rather unexpectedmolecular heterogeneity of Golgi neurons themselves (e.g.,Geurts et al. 2003; Simat et al. 2007), that waits to beunderstood functionally.

Classic Lugaro cells have a fusiform cell body and areintermediate in size between granule cells and Golgi cells.Consequently, they are often referred to also as fusiformcells of Lugaro. They are found in all parts of the cerebellarcortex, where they are located at the border between thePurkinje cell layer and the upper part of the granule celllayer. Initial numerical estimates had put the Purkinje cell/Lugaro cell ratio at about 15:1 in the rat (Dieudonne andDumoulin 2000) and 30:1 in the cat (Sahin and HockWeld1990). More recently, a detailed neurochemical analysis oflarge granule cell layer interneurons indicated that in themouse, Lugaro cells account for about 1/3 of all inhibitorygranule cell layer interneurons, and thus are more numerousthan initially estimated (Simat et al. 2007).

The longer axis of classical, fusiform cells of Lugaro isoriented in the parasagittal plane. Typically, it emanatestwo pairs of long, rarely dividing dendrites, which run justunderneath the Purkinje cell layer, also in the parasagittalplane. However, as the two dendrites originating from onepole of the cell body also diverge from each other in thehorizontal plane, the actual dendritic Weld covered by a typ-ical Lugaro cell may be better perceived as an rectangle,located just underneath the Purkinje cell layer, with its longaxis parallel to the sagittal plane. These dendrites are ofrather variable length, measuring, in the rat, from some 100to 700 �m (Laine and Axelrad 1996). Lugaro cells are cur-rently viewed as the primary target of serotoninergic inputinto the cerebellar cortex (Dieudonne and Dumoulin 2000).They are also innervated by recurrent Purkinje cell axoncollaterals (Palay and Chan-Palay 1974), and basket cells(Fox et al. 1967). Finally, it has been observed that Lugarocells are sensitive to ADP, as mediated through P2Y puri-noreceptors, and that P2Y receptor activation may modu-late inhibitory input from Lugaro to Purkinje cells (seebelow) in a complex spatio-temporal pattern (Saitow et al.2005). The most likely source of ATP/ADP seem to be cer-ebellar astrocytes, in particular Bergmann glia (Saitowet al. 2005).

While the actual course of Lugaro cell axons may bequite variable, their deWning feature is that they all terminate

123


within the molecular layer, although they also give oV col-laterals into the granule cell layer (Laine and Axelrad1996). All Lugaro cells have been reported to have a localaxonal projection into the molecular layer just above theoriginating cell perikaryon; in addition, some Lugaro cellsalso have projections that reach more distal targets, which,however, are again located in the molecular layer. On theirway, these longer axons may even course through the whitematter (Laine and Axelrad 1996). As direct targets of Lug-aro cells, basket and stellate cells have been identiWed onlyby morphological means (Laine and Axelrad 1998),whereas the projections of Lugaro cells onto Golgi (Die-udonne and Dumoulin 2000; Dumoulin et al. 2001; Die-udonne 1995) and Purkinje cells (Dean et al. 2003) havebeen analyzed functionally. It has been estimated that oneLugaro cell projects to more than 100 Golgi cells (Die-udonne and Dumoulin 2000); comparable numerical esti-mates for the Lugaro to Purkinje cell, or Lugaro to basket/stellate cell projections are not available.

Whereas the Lugaro cell input to Golgi cells has mixedGABAergic and glycinergic components (Dumoulin et al.2001), their input to (juvenile; analyzed at postnatal day 14in the rat) Purkinje cells is mediated only by GABAA recep-tors (Dean et al. 2003). (Adult) Purkinje cells do expressglycine receptors (Triller et al. 1987), though these arerather sparsely localized on main dendritic shafts. It is pres-ently not clear whether the failure to detect glycinergicinput into Purkinje cells by Lugaro cells relates to thedevelopmental expression of glycinergic receptors by Pur-kinje cells, or rather reXects diVerential loading of Lugarocell vesicles targeted to Lugaro/Golgi and Lugaro/Purkinjecell synapses, respectively (for a model, cf Fremeau et al.2004; Schuske and Jorgensen 2004). As the vesicular trans-porter VGAT/VIAAT is nonselective for GABA or glycine(which compete for it; Chaudhry et al. 1998; Wojcik et al.2006), such a scenario would imply diVerential sorting and/or membrane availability of membrane transporters for gly-cine and GABA. Indeed, there is a growing body of evi-dence that points to an intricate regulation of the sortingand membrane insertion of Glyt2, Gat-1, and Gat-2 (e.g.,Muth et al. 1998; Martinez-Maza et al. 2001; Farhan et al.2008; see also Chiu et al. 2002), which are expressed inlarge molecular layer interneurons of the cerebellum (cf theAllen Brain Atlas; Lein et al. 2007). Intriguingly, Gat-2 issubject to regulation by serotonin; as mentioned above,Lugaro cells are the primary target of serotoninergic projec-tions to the cerebellar cortex (Dieudonne and Dumoulin2000).

Besides classical, spindle-shaped Lugaro cells in theupper part of the granule cell layer, a second type of neuronis now grouped as Lugaro cells, primarily based on its typi-cal wiring and axonal projection pattern (Fig. 2). Thesecells were recently described by Laine and Axelrad (2002).

They share with the classical Lugaro cells just described thetypical projection of their axons; yet they occupy a deeperposition within the granule cell layer, and they have a pref-erentially globular perikaryon. Also, the expanse and pat-tern of their dendritic processes are somewhat reminiscentof Golgi cells. In addition to their axonal projection, immu-noreactivity for calretinin and their innervation by recurrentPurkinje cell axons provide convincing arguments to clas-sify these cells indeed as Lugaro cells (Laine and Axelrad2002). Apparently these cells are identical to “Golgi-like”calretinin-positive but mGluR2-negative cells described byGeurts et al. (2001), although the two reports diVer as towhether these cells are preferentially found in the moresuperWcial (Laine and Axelrad 2002)or deeper parts (Geurtset al. 2001) of the granule cell layer.

Of all cerebellar cortical neurons, Lugaro cells are theonly ones stained by monoclonal antibody Cat-301 (Sahinand HockWeld 1990), although it is not clear whether allLugaro cells are positive for this marker. Outside the cere-bellar cortex, this antibody recognizes multiple types ofneurons, and its antigen is subject to developmental andactivity-dependent regulation (Lander et al. 1997). Molecu-larly, it has been identiWed as a distinctly glycosylated formof the extracellular matrix protein, aggrecan, which formspart of perineuronal nets (Lander et al. 1998; Matthewset al. 2002). Cat-301 is believed to have a role in the stabil-ization of synaptic structures, although no deWnitive role forthis antigen has been identiWed.

There is some indication, though no deWnitive data, thatLugaro cells start to be born at embryonic day 14–15, in therat (Sekerkova et al. 2004), and in all likelihood they arise(together with) other inhibitory interneurons of the cerebel-lar cortex, from the fourth ventricle neuroepithelium.

As a group, Golgi cells diVer from Lugaro cells primar-ily by their axonal projection pattern: While Lugaro cells,including their globular variant, target all other inhibitoryinterneurons of the cerebellar cortex, Golgi cells projectselectively to granule cells and UBCs, which they contactwithin mossy Wber glomeruli. As Golgi cells receive bothmossy Wber and granule cell (parallel Wber) input, they arepoised to realize both a feed-forward and a feed-back looponto granule cells. It has been known since the days ofRamon y Cajal that Golgi cells show quite a degree of het-erogeneity, and Cajal initially proposed the existence offour variants, based on axonal projection patterns (Ramon yCajal 1899). Since then, various classiWcation schemeshave been proposed, of which the most recent one, bySimat et al. (2007), incorporates molecular data that holdsthe prospect for deWning functional classiWcations. Theseauthors provide evidence that allows the delineation of Wvetypes of Golgi cells, based on the variable expression of theGABA-synthetic enzyme, Gad67, the cell membrane trans-porter for glycine, GlyT2, of secretogranin, and of the

123


metabotropic glutamate receptor mGluR2. About two-thirds of all Golgi cells, referred to as type 1, are positivefor Gad67, GlyT2 (and hence GABAergic and glycinergic),secretogranin, and also mGluR2. A second set (type 2),comprising less than 10% of all Golgi cells, shares the samebasic neurochemical makeup but does not express neurogr-anin. Type 3 Golgi cells encompass actually two subsets,each comprising some 2–5% of all Golgi cells, that shareexpression of mGlutR2 and GlyT2, but diVer as only onesubset expresses Gad67. Type 4 cells are pure GABAergiccells that do not express GlyT2, and are also negative formGluR2; they are, however, immunopositive for neurogra-nin. These type 4 cells comprise about 15% of all Golgicells in mice. Finally, less than 5% of all Golgi cells are oftype 5, which does not express Gad67, nor mGluR2, norneurogranin, but is positive for GlyT2. These neurochemi-cal markers can also be utilized to tell apart Golgi cellsfrom Lugaro cells, including their globular variant: Lugarocells are all positive for both GlyT2 and GAD67, but nega-tive for mGluR2,3 and neurogranin.

So far, these neurochemical diVerences have not beenconsidered in any model trying to describe Golgi and/orLugaro cell function, although the need to do so was clearlystressed already some 35 years ago (Mugnaini 1972). Onecaveat to be heeded in any attempt to integrate the excitingdata of Simat et al. (2007) in such a scheme is that theseauthors relied on GFP expression from the Gad67 (Gad1)and Glyt2 locus. It has not been formally shown that thesetransgenes represent expression of cognate genes in all celltypes analyzed, although available evidence leaves littledoubt that this is indeed so. While coexpression of GAB-Aergic and glycinergic traits has also been observed in rats(Tanaka and Ezure 2004) and primates (Ottersen et al.1987), exact numerical data are missing, so that we cannoteven speculate about possible species diVerences, and theevolutionary signiWcance.

A critical issue when trying to assemble these recent datain a functional model is that we have no real idea whetherthe expression of genes used to deWne Golgi and also Lug-aro cell subsets is indeed stable, or subject to dynamic, orplastic, regulation. This is particularly relevant in view ofrecent data that suggest that (terminal?) diVerentiation ofcerebellar cortical inhibitory interneurons, including Golgiand Lugaro cells, but also basket and stellate neurons, issubstantially driven by local cues impinging on pluripotentprecursors of these cells (Leto et al. 2006) (see below).

Inhibitory interneurons of the molecular layer: basket and stellate cells, or basket/stellate cells?

As their names imply, basket and stellate cells were ini-tially discerned based on the typical morphology of their

axon (terminals), or their dendrites, respectively, and alsoby the position of their perikarya in the lower or uppermolecular layer (Ramon y Cajal 1909). A systematic analy-sis of an admittedly small set of molecular layer neurons(but still the largest set that was systematically analyzed)indeed conWrmed that the tendency to form a basket cell-type of axon terminal in fact paralleled the position of theperikaryon within the molecular layer (Sultan and Bower1998), as has oft been suggested based on observations inGolgi-stained material. Importantly, however, this studyalso suggested that molecular layer interneurons, if classi-Wed based on morphological criteria, constitute a singlepopulation that varies only gradually. Clearly, follow up onthis notion must keep in mind the small experimental setavailable for this systematic analysis.

A molecular handle to address this issue was suggestedby the observation that in mice null for cyclin D2, neuronsfrom the upper molecular layer are conspicuously lacking.This has led to the suggestion that cyclin D2 is a key deter-minant of stellate cell formation, and indeed is required forthe developmental “appearance of this sublineage” (Huardet al. 1999): This notion seems also supported by the obser-vation that in the forebrain of cyclin D2-deWcient mice, dis-crete subsets of interneurons are missing (Glickstein et al.2007). However, this cannot be interpreted as an indicationthat cyclin D2 acts to “specify” a stellate cell-speciWc line-age. As pointed out (Huard et al. 1999; Glickstein et al.2007), lack of cyclin D2 impedes proliferation, and it iswell conceivable that this may lead to an exhaustion of theprecursor cell pool from which cerebellar inhibitory inter-neurons originate before stellate cells—which are known tobe born last—are indeed formed. Such an interpretationwould be in keeping with the data that document that termi-nal diVerentiation, including the acquisition and expressionof cell type speciWc traits for cerebellar inhibitory interneu-rons is driven by local cues. Intriguingly, cyclin D2 mayplay a more active role in this scenario than just supportinga suYcient number of cell cycles needed to generate a fullcomplement of inhibitory cerebellar interneurons (seebelow). Other than cyclin D2, no molecular cues to diVer-entiate basket from stellate cells have emerged so far.

Development and diVerentiation of cerebellar (cortical) inhibitory interneurons

While there is some controversy in the older literature as tothe site of origin of cerebellar inhibitory interneurons, thehistorical consensus, or at least the prevailing view was thatbasket and stellate cells are derived from the external gran-ule cell layer, and Golgi cells from the ventricular epithe-lium layer (see, e.g., Altman and Bayer 1997). Indirectly,this suggested the notion that the lineages of these cell

123


types are well separated, in space, and presumably also indevelopmental potential. Although a series of earlier obser-vations had put this view in question (prominent amongthem (Napieralski and Eisenman 1993; Hallonet and LeDouarin 1993) the view that all inhibitory cortical interneu-rons derive from a precursor population—and maybe evena common precursor—that reaches its Wnal destinations bymigrating through the deep cerebellar mass and nascentwhite matter gained broader attention and general accep-tance following the reports by Zhang and Goldman (1996a,b) that inhibitory interneuronal precursors could be markedby injecting a retroviral marker into the deep parts of thecerebellar anlage. These studies also implied that precur-sors in this location (i.e., which had left the ventricularzone) were dividing, and they also raised the issue of apotential lineage relationship between inhibitory interneu-ronal precursors and other cells marked, notably glial cells(see also Mathis et al. 1997). A signiWcant advance, and amethodological cornerstone for further research, wasreached when Maricich and Herrup (1999) could deWnePax2 as a marker for inhibitory interneuronal precursors.This allowed, for the Wrst time, to localize these cellsthroughout cerebellar development, and to infer the paththat they take from the ventricular epithelium through thedeep cerebellar mass and nascent white matter into the cer-ebellar cortex (Fig. 3). The assertion of Maricich and Her-rup (1999) that Pax2 positive interneuronal precursors forma neuronogenic population distinct from glial precursorswas further substantiated by data which documented thatdividing precursors in the deep cerebellar mass of 4–5 dayold rat pups yield, upon further diVerentiation, descendants/

clones which contained either cells with neuronal, astrog-lial, or oligodendroglial molecular signatures, but only veryrarely mixed clones (Milosevic and Goldman 2002). Fur-ther in vitro studies by these authors (Milosevic and Gold-man 2004) showed that mixed clones were obtained morefrequently from proliferating precursors isolated from thecerebellar ventricular neuroepithelium, but only rarely fromproliferating precursors isolated from the nascent whitematter, implying at least some diVerentiation and lineagerestriction of proliferating precursor cells migrating into thecerebellar anlage. Moreover, a quantitative analysis ofPax2-positive cells in the cerebellar anlage documentedthat only a minor fraction of these cells proliferate, and thatindeed the numerical increase seen in Pax2-positive cere-bellar cortical inhibitory interneurons in the early postnatalphase (assessed at postnatal days 0 and 3 in the mouse) canonly be explained by proliferation of a Pax2-negative pre-cursor population, and that Pax2-expression commencesclose to the time point that these cells go through their lastmitosis (Weisheit et al. 2006). Together, these data suggestthat Pax2 should be perceived as a marker for early diVer-entiating interneuronal precursors. Finally, it is only afterthey leave the ventricular epithelium that inhibitory inter-neuronal precursors become positive for Pax2 (Maricichand Herrup 1999); cf also Vilz et al. 2005 and Zordan et al.2008) which again is in keeping with the observation ofHerrup and associates (Maricich and Herrup 1999) thatPax2-positive cells, in contrast to ventricular epithelia, arestrictly neuronogenic.

The notion that all GABAergic cerebellar interneuronsshare a common, molecularly deWned lineage, as implied

Fig. 3 a Location and morphology of Pax2-GFP-positive precursorsof inhibitory interneurons (green) in the cerebellar cortex of an 8-dayold mouse. Cell nuclei are counterstained in red. This image representsan optical section of 3.5 �m thickness. b Same area as shown in panelA, but scanned to a tissue thickness of a total of 50 �m. Only the Pax2-GFP signal is shown, and the position of positive cells along the z axis(depth in the tissue block) is color-coded to give an impression of the

three-dimensional distribution of Pax2-positive interneuronal precur-sors. Note diVerences in cell shape, orientation and size in the nascentwhite matter (wm), internal granule cell layer (igl) and molecular layer.Note also the accumulation of Pax2-GFP-positive cells underneath theexternal granule cell layer (egl), which is not penetrated. p Purkinje celllayer. Bar 50 �m

123


by these observations, is further supported by data thatshowed that genesis of these neurons, including that ofGABAergic Purkinje cells, is critically dependent on Ptf1aexpression (Hoshino et al. 2005). In contrast, initial forma-tion of granule cell precursors in the nascent external gran-ule cell layer is not aVected by mutating this gene actually,as elegantly shown by Pascual et al. (2007) lack of Ptf1adoes not result in a failure of GABAergic precursors toform, or to their demise, but redirects these cells to acquirean excitatory, external granule cell layer fate. The fate anddevelopment of (purely) glycinergic cerebellar neurons(i.e., type 5 Golgi cells according to Simat et al. (2007) hasnot been analyzed in Ptf1a-null mice. In the retina, Ptf1a isalso critical for the development of glycinergic cells (Nak-hai et al. 2007). Therefore, it is tempting to speculate thatPtf1a is critical to development of all inhibitory neurons ofthe cerebellum.

In the forebrain, GABAergic interneuron diversity of theadult cortex is preceded by, and implemented through, spa-tial genetic patterning of the embryonic neuroepithelium(e.g., Fogarty et al. 2007). Analysis of proneural geneexpression in the cerebellar anlage revealed that the earlyembryonic ventricular epithelium, from which Purkinjecells, and inhibitory cerebellar interneurons arise, is indeedheterogeneous, as documented by a patchy, and only partlyoverlapping pattern of expression of Mash1, Neurog1, andNeurog2 between embryonic day 10.75 and 13.5 (Zordanet al. 2008). Yet whether, and how these so deWned popula-tions relate to adult subsets of inhibitory cerebellar inter-neurons is not yet known.

Indeed, there is compelling evidence that inhibitoryinterneuronal precursor, rather than being irrevocably fatedand determined within (or close to) the ventricular neuroep-ithelium, maintain a considerable level of developmentalplasticity as they translocate through the nascent cerebellaranlage, and that their eventual fate and diVerentiation into,say molecular layer interneurons or Golgi cells is deter-mined by local cues in their eventual territory of residency.This model is based on a large and systematic analysis ofthe development of genetically tagged cerebellar precursorsfollowing heterotopic and heterochronic transplantationinto cerebella of embryonic, early postnatal, and adult hosts(Leto et al. 2006; see also Leto et al. 2008). Importantly,these studies also showed that the developmental potentialof interneuronal precursors is not progressively restrictedover the time course of cerebellar histogenesis. Thus, evenprecursors that are normally destined to form stellate cells(i.e., the type of inhibitory interneuron formed last duringnormal development) can give rise to, say, Golgi cells, oreven deep nuclear inhibitory interneurons, which normallyare formed and diVerentiate much earlier, when trans-planted into the (internal) granule cell layer, the normalGolgi “residential area”, or into the region of deep nuclei.

This protracted developmental plasticity, which persistswell beyond the terminal mitosis of cerebellar inhibitoryinterneuronal precursors, clearly distinguishes them, say,from forebrain neurons. The developmental potential offorebrain cortical neuronal precursors becomes progres-sively restricted over time (Desai and McConnell 2000),and these cells are typically fated by signals received dur-ing a rather restricted time window around their last mito-sis, within or close to the ventricular zone. Another keyfeature of cortical neuron development is that such fate-deWning signals can only be realized by cycling cells, andindeed cell sensitivity toward such signals varies over thecell cycle (e.g., McConnell and Kaznowski 1991; Fukumi-tsu et al. 2006; Leone et al. 2008).

Yet there is also evidence suggesting that these seem-ingly so diverse time courses and strategies of neuronal fatedetermination might actually be subserved by the same, orclosely related (molecular) mechanisms. A Wrst hint camefrom in vitro studies that documented that development ofcerebellar inhibitory interneuronal precursors, like that ofgranule cells, is indeed sensitive to extrinsic signals in acell cycle dependent way (Baader et al. 1999). Moreover, ithas been realized that migrating, Pax2-positive inhibitoryinterneuronal precursors in the cerebellar anlage maintainexpression of the proliferation marker, Ki67, although theydo not actually divide. Thus, in precursors of basket andstellate cells, Ki67 expression can be observed up to theirarrival in the molecular layer (Weisheit et al. 2006). AndWnally, the above mentioned dependence of stellate celldevelopment on cyclin D2 (Huard et al. 1999) needs to bereconsidered in this context. In the cerebellar anlage, cyclinD2 is expressed in the external granule cell layer and inwhat seem to be largely inhibitory interneuronal precursorstransiting through the nascent white matter and forminglayers of the cerebellum (cf Fig. 6a of Ciemerych et al.2002; see also the cyclin D2 expression pattern at postnatalday 7 as documented in the BGEM [Magdaleno et al. 2006)and genepaint (Visel et al. 2004) databases]. Beyond itsrole in cell cycle progression, cyclin D2 can act to induceand/or maintain a non-proliferative state (Meyyappan et al.1998); thus it is tempting to speculate whether sustainedexpression of cyclin D2 primarily restricts proliferativeactivity of translocating precursors of cerebellar inhibitoryinterneurons. Indeed, Huard et al. (1999) already suggestedthat the lack of stellate cells in cyclin D2-null mice may beexplained by eVects other than that on cell cycle progression.

In any case, the observation of cyclin D2 and Ki67expression in migratory, non-dividing precursors of cere-bellar inhibitory interneurons document that these cells per-sist in a state clearly distinct from full-blown G0 phase.Thus, a model may be envisaged according to which theterminal fate and diVerentiation of cerebellar inhibitoryinterneurons, like that of many other central nervous

123


neurons, is deWned by cues acting during a phase related totheir last mitosis. What distinguishes cerebellar inhibitoryinterneurons form, say, forebrain cortical projection neu-rons, is that this phase is apparently extended, as revealedby (and caused/implemented through?) their continuedexpression of cyclin D2 and Ki67. As a consequence, thesecells are capable of integrating environmental developmen-tal signals far oV the immediate vicinity of their Wnal mito-sis, in areas they reach only after protracted translocation.The observation that mice null for the ErbB4 receptor haveincreased numbers of large granule cell layer inhibitoryinterneurons suggests a role for the ErbB4/neuregulin sig-naling pathway in this scenario (Tidcombe et al. 2003).Further, in vitro observations support the notion that suchlocal cues include electrical activity and BDNF, which havebeen noted to regulate the expression of mGluR2 and par-valbumin, and also dendritic morphogenesis of cerebellarinterneurons (Koscheck et al. 2003; Mertz et al. 2000).These data suggest that the interpretation of such signals,and speciWcally that of BDNF, might be critically modu-lated either by the electrical activity, or geometricalconstraints, similar to those that migrating interneuronal pre-cursors encounter in various parts of the cerebellar anlagethey have to traverse, or where they eventually settle.

Migration of cerebellar inhibitory interneurons and neuritogenesis

A question that immediately follows, then, is: how ismigration of cerebellar inhibitory interneuronal precursorsregulated? Mathis et al. suggested a role for oligodendroy-tes and/or their precursors, based on the observation thattransgenic ablation of these cells results in a severe distur-bance of the migration and/or diVerentiation of cerebellarcortical inhibitory interneurons (Mathis et al. 2003). How-ever, the thymidine kinase/FIAU system they used to elimi-nate oligodendritic cells is notorious for its ability to killnot only those cells primarily targeted, but also nearbycells. Indeed, this “bystander”-eVect is currently activelyexploited for the development of stem-cell based anti-tumor therapies (e.g., Uhl et al. 2005; Li et al. 2005). ThiseVect is facilitated by gap-junctional coupling (Mesnil andYamasaki 2000), which may be surmised to be active alsoin Pax2-positive precursors of cerebellar cortical inhibitoryinterneurons, that, like cerebellar oligodendrocytic cells,express connexins (Maxeiner et al. 2003; Kunzelmannet al. 1997). Thus, the physiological signiWcance of theobservations of Mathis et al. (2003) is currently hard tofathom.

A better and more in-depth understanding of the mecha-nisms that regulate the migration and appropriate distribu-tion of cerebellar interneurons seems of prime importance,

given that the Wnal diVerentiation and function of thesecells is eventually triggered in their area of residency (cfabove; see Leto et al. 2006). Further, as there occurs butvery little apoptosis, at least among basket and stellate neu-rons (Yamanaka et al. 2004), cell elimination seems ofminor signiWcance to assure numerical matching and appro-priate spacing. Migrating interneuronal precursors are con-fronted with structurally quite diVerent territories, includingthe nascent white matter, the internal granule cell layer,and, after becoming molecular layer interneurons, also thePurkinje and molecular layers. Conceptually, the issue ofhow proper navigation and settling is regulated may be bro-ken down into several broad questions. E.g., what triggersinterneuronal precursors migrating through the nascentwhite matter to leave this territory either at the base, or theapex of an emerging cerebellar folium? How is the decisionto either stop in the nascent (internal) granule cell layer, orto proceed in to the molecular layer implemented in molec-ular terms? Do interneurons that have reached their Wnaldestination signal-back to their brethren still en route andthereby impinge on the latter’s migrational behaviour, e.g.,by modulation of Purkinje cell output which would then besensed by migratory cells in the nascent white matter?Functional integration of both Golgi and basket/stellatecells (i.e., inhibiton of granule cells and Purkinje cells,respectively), which would be a prerequisite for such a sce-nario, has been documented, in the rat, at about postnataldays 10–12, (Shimono et al. 1976), i.e. at about midwayduring the formation of the molecular and granule cell lay-ers. Once inhibitory interneurons have reached their lami-nar destination, how is their appropriate spacing achieved?And, related to this, how is the tiling of their dendrites, andthe precise targeting of their axons realized? We have noreal answers to the Wrst three of these questions, but modelsystems to address these issues are now about to be devel-oped (see, e.g., Hecker et al. 2008).

In contrast, recent results suggest at least some basicprinciples governing basket/stellate cell axon formation.Cerebellar Purkinje cells have been historically an out-standing example to study subcellular segregation ofdiverse aVerents and their (mutual) regulation, includingparallel and climbing Wbers (e.g., Chen and Hillman 1982;Sotelo 1990; Cesa et al. 2007), both of which synapse ontodiscernible types of dendritic spines. Basket cells project tothe Purkinje cell bodies and the axon initial hillock actuallythis is the very prerequisite to deWne a basket cell. Stellatecells in contrast, project to more distant dendritic shafts.This diVerential innervation by inhibitory interneurons isparalleled, and may indeed be realized through a gradient ofcell adhesivity mediated by the subcellular distribution ofneurofascin (Ango et al. 2004). A parsimonious explana-tion then would be that the decision to develop into a basketor stellate cell would eventually reXect the timing of when

123


the axons of these cells actually invest Purkinje cells. Thosethat arrive early on can secure highly attractive (and adhe-sive) Purkinje cell territories around the axon hillock andcell body; axons which reach Purkinje cells later on have tosettle for less adhesive distal dendritic shafts. As thosemolecular layer interneurons which arrive Wrst in themolecular layer, and presumably also have a head-start toelaborate their axons, settle in the lower molecular layer(Miale and Sidman 1961; Altman and Bayer 1997), thiswould also explain the general conception that basket cells(preferentially) reside in the lower molecular layer. Cellsthat arrive later in the molecular layer and settle moresuperWcially, and would Wnd only more distal dendrites freeto innervate. A twist to this story is added by the Wndingthat developmental interaction between Purkinje cells andtheir aVerent inhibitory neurons is mediated, or at leastmodulated, by Bergmann astroglial cells (Ango et al.2008).

More markers, more cells?

Localization of an ever expanding panel of moleculeswithin the cerebellar cortex may be expected to necessitateongoing modiWcation of the traditional classiWcations ofcerebellar interneurons, but also to eventually facilitate ourunderstanding of the development and function of thesecells. An example in question is the novel neurochemicaldiversity of Golgi cells (Simat et al. 2007) sketched above.Meanwhile, histological and immunocytochemical analysisof the cerebellum has also suggested the potential existenceof additional types of neurons. Thus, Crook et al. havedescribed an apparently very rare, but rather large neuronalphenotype localized within the cerebellar white matter(Crook et al. 2007). These cat-301 positive cells are con-tacted (innervated) by GAD/calbindin D28 k-doubly immu-noreactive collaterals (i.e., Purkinje cells). Their exactwiring is elusive, as is their function. Interestingly, anapparently very similar, if not identical cellular phenotypehas also been observed in the cerebellum of several aquaticmammals, as reported by Addison already in 1931 (Addi-son 1931). Further, these cells are also reminiscent of thesynarmotic cells described Wrst by Landau (1927) and laterby Braak (1974), who also argued that they might be a sub-set of Lugaro cells.

Staining of murine cerebella for Npas3 highlighted agroup of cells positive for this basic HLH transcription fac-tor, located within the granule cell layer, the numbers anddistribution of which does not conform with any knowncerebellar cell type (Erbel-Sieler et al. 2004). Npas3-nullmice show a behavioral phenotype pointing to a cerebellardysfunction (Brunskill et al. 2005). Still, the function andnature of these cells remains enigmatic.

Finally, staining of murine cerebella for NeuN high-lighted a group of cells, located in the lower molecularlayer, which so far have not been identiWed (Weyer andSchilling 2003). This study also showed that molecularlyidentiWed cerebellar cortical inhibitory interneurons do notstain for NeuN, just like Purkinje cells, and also unipolarbrush cells. Among molecular markers discussed here,NeuN is the exception in that it is not genetically deWned(see also Lind et al. 2004). For all other markers, the abilityto eventually tag the cellular phenotypes expressing themvitally will not only allow to follow these cells duringdevelopment, but also to identify them in vivo for physio-logical studies.

Concluding remarks

While the results summarized above leave little doubt thatcerebellar inhibitory interneurons constitute a more diverseand complex complement of cells than traditional classiW-cations acknowledge, these results also provide vantagepoints that should eventually allow us to arrive at an inte-grated view of how the cerebellar circuitry is formed andfunctions. Needless to say, this goal is still far oV, and as isso often the case, novel results have also generated novelquestions. Fortunately, recent progress has not only focusedattention on cerebellar interneurons, but also providesimportant tools and approaches to follow up on these leads.Clearly, we would like to learn in much more detail howinterneuronal precursors navigate the cerebellar anlage, andhow they eventually become integrated in the cerebellarcircuitry. Also, does the molecular diversity that has beenuncovered, and that certainly will still be expanded, reXectthe endpoint of a developmental process, or rather an aspectof functional plasticity? A closely related question, of bothdevelopmental and clinical interest, is whether there existsany relationship between inhibitory interneuronal lineagesand cerebellar tumorigenesis. To date, the developmentalorigin of only a subgroup of medulloblastoma, the mostcommon and devastating cerebellar malignant tumor, couldbe traced to the granule cell lineage (cf Pietsch et al. 2004;Gilbertson and Ellison 2008 for detailed discussions andprimary references). It is therefore tempting to speculatewhether other forms of medulloblastoma might be derivedfrom cells that normally would form inhibitory cerebellar(inter) neurons. Currently, there is hardly any evidence tothis end, but one has to consider that we still are ratherignorant with respect to molecular markers that mightdeWne the inhibitory interneuronal lineage. Pax2 isexpressed in it only around its last mitosis (cf above; seealso (Weisheit et al. 2006), and thus it may not be surpris-ing that this relatively “late” marker was not found inmedulloblastoma (Kozmik et al. 1995).

123


Clearly, the realization that cerebellar circuitry com-prises a more diverse set of cellular constituents than tradi-tionally perceived has made it a more complex, but also amore interesting paradigm to analyze the mechanisms thatbring about the formation and proper function of the centralnervous system. While we are faced with the challenge yetto integrate these diverse cellular phenotypes into a coher-ent picture of how the cerebellum works, it may be hopedthat the appreciation and clariWcation of this diversity willalso provide an opportunity to overcome some of the road-blocks that still hamper progress towards a coherent viewof cerebellar function.

Acknowledgments We thank various members of our labs for help-ful discussions and suggestions. Work in our laboratories was sup-ported by grants from the DAAD, NSF grant #IBN-0138147 to J.O.,the Compagnia di San Paolo (Neurotransplant Project, 2004.2019 –2007.0660), and the Regione Piemonte (Proj. A14/05 and 865/2006).

Open Access This article is distributed under the terms of the Crea-tive Commons Attribution Noncommercial License which permits anynoncommercial use, distribution, and reproduction in any medium,provided the original author(s) and source are credited.

References

Addison WHF (1931) Unusual large nerve cells in the cerebellar cortexof several aquatic mammals. Anat Rec 48:7–72

Altman J, Bayer SA (1977) Time of origin and distribution of a newcell type in the rat cerebellar cortex. Exp Brain Res 29:265–274

Altman J, Bayer SA (1997) Development of the cerebellar system inrelation to its evolution, structure, and functions. CRC Press,Boca Raton

Anelli R, Dunn ME, Mugnaini E (2000) Unipolar brush cells developa set of characteristic features in primary cerebellar cultures. JNeurocytol 29:129–144

Anelli R, Mugnaini E (2001) Enrichment of unipolar brush cell-likeneurons in primary rat cerebellar cultures. Anat Embryol (Berlin)203:283–292

Ango F, Di CG, Higashiyama H, Bennett V, Wu P, Huang ZJ (2004)Ankyrin-based subcellular gradient of neurofascin, an immuno-globulin family protein, directs GABAergic innervation at pur-kinje axon initial segment. Cell 119:257–272

Ango F, Wu C, van der Want JJ, Wu P, Schachner M, Huang ZJ (2008)Bergmann glia and the recognition molecule CHL1 organizeGABAergic axons and direct innervation of Purkinje cell den-drites. PLoS Biol 6:e103

Apps R, Garwicz M (2005) Anatomical and physiological foundationsof cerebellar information processing. Nat Rev Neurosci 6:297–311

Baader SL, Bergmann M, Mertz K, Fox PA, Gerdes J, Oberdick J,Schilling K (1999) The diVerentiation of cerebellar interneurons isindependent of their mitotic history. Neuroscience 90:1243–1254

Berthie B, Axelrad H (1994) Granular layer collaterals of the unipolarbrush cell axon display rosette-like excrescences. A Golgi studyin the rat cerebellar cortex. Neurosci Lett 167:161–165

Boyden ES, Katoh A, Raymond JL (2004) Cerebellum-dependentlearning: the role of multiple plasticity mechanisms. Annu RevNeurosci. 27:581–609

Braak H (1974) On the intermediate cells of lugaro within the cerebel-lar cortex of man. A pigmentarchitectonic study. Cell Tissue Res149:399–411

Brockhaus J, Dressel D, Herold S, Deitmer JW (2004) Purinergic mod-ulation of synaptic input to Purkinje neurons in rat cerebellarbrain slices. Eur J NeuroSci 19:2221–2230

Brunskill EW, Ehrman LA, Williams MT, Klanke J, Hammer D,Schaefer TL, Sah R, Dorn GW, Potter SS, Vorhees CV (2005)Abnormal neurodevelopment, neurosignaling and behaviour inNpas3-deWcient mice. Eur J NeuroSci 22:1265–1276

Caviness VS Jr, Rakic P (1976) Mechanisms of cortical development:a view from mutant mice. Annu Rev Neurosci 1:297–326

Cesa R, Scelfo B, Strata P (2007) Activity-dependent presynaptic andpostsynaptic structural plasticity in the mature cerebellum. J Neu-rosci 27:4603–4611

Chaudhry FA, Reimer RJ, Bellocchio EE, Danbolt NC, Osen KK, Ed-wards RH, Storm-Mathisen J (1998) The vesicular GABA trans-porter, VGAT, localizes to synaptic vesicles in sets of glycinergicas well as GABAergic neurons. J Neurosci 18:9733–9750

Chen S, Hillman DE (1982) Plasticity of the parallel Wber-Purkinje cellsynapse by spine takeover and new synapse formation in the adultrat. Brain Res 240:205–220

Chiu CS, Jensen K, Sokolova I, Wang D, Li M, Deshpande P, David-son N, Mody I, Quick MW, Quake SR, Lester HA (2002) Num-ber, density, and surface/cytoplasmic distribution of GABAtransporters at presynaptic structures of knock-in mice carryingGABA transporter subtype 1-green Xuorescent protein fusions. JNeurosci 22:10251–10266

Ciemerych MA, Kenney AM, Sicinska E, Kalaszczynska I, BronsonRT, Rowitch DH, Gardner H, Sicinski P (2002) Development ofmice expressing a single D-type cyclin. Genes Dev 16:3277–3289

Crook J, Hendrickson A, Robinson FR (2006) Co-localization of gly-cine and gaba immunoreactivity in interneurons in Macaca mon-key cerebellar cortex. Neuroscience 141:1951–1959

Crook JD, Hendrickson A, Erickson A, Possin D, Robinson FR (2007)Purkinje cell axon collaterals terminate on Cat-301+ neurons inMacaca monkey cerebellum. Neuroscience 149:834–844

Dean I, Robertson SJ, Edwards FA (2003) Serotonin drives a novel GAB-Aergic synaptic current recorded in rat cerebellar purkinje cells: aLugaro cell to Purkinje cell synapse. J Neurosci 23:4457–4469

Desai AR, McConnell SK (2000) Progressive restriction in fate poten-tial by neural progenitors during cerebral cortical Development.Development 127:2863–2872

Dieudonne S (1995) Glycinergic synaptic currents in Golgi cells of therat cerebellum. Proc Natl Acad Sci USA 92:1441–1445

Dieudonne S, Dumoulin A (2000) Serotonin-driven long-range inhibi-tory connections in the cerebellar cortex. J Neurosci 20:1837–1848

Dino MR, Schuerger RJ, Liu Y, Slater NT, Mugnaini E (2000) Unipo-lar brush cell: a potential feedforward excitatory interneuron ofthe cerebellum. Neuroscience 98:625–636

Dugue GP, Dumoulin A, Triller A, Dieudonne S (2005) Target-depen-dent use of coreleased inhibitory transmitters at central synapses.J Neurosci 25:6490–6498

Dumoulin A, Triller A, Dieudonne S (2001) IPSC kinetics at identiWedGABAergic and mixed GABAergic and glycinergic synapsesonto cerebellar Golgi cells. J Neurosci 21:6045–6057

Englund C, Kowalczyk T, Daza RA, Dagan A, Lau C, Rose MF, Hev-ner RF (2006) Unipolar brush cells of the cerebellum are pro-duced in the rhombic lip and migrate through developing whitematter. J Neurosci 26:9184–9195

Erbel-Sieler C, Dudley C, Zhou Y, Wu X, Estill SJ, Han T, Diaz-Ar-rastia R, Brunskill EW, Potter SS, McKnight SL (2004) Behav-ioral and regulatory abnormalities in mice deWcient in the NPAS1and NPAS3 transcription factors. Proc Natl Acad Sci USA101:13648–13653

Farhan H, Reiterer V, Kriz A, Hauri HP, Pavelka M, Sitte HH, Freiss-muth M (2008) Signal-dependent export of GABA transporter 1from the ER-Golgi intermediate compartment is speciWed by a C-terminal motif. J Cell Sci 121:753–761

123


Flace P, Benagiano V, Lorusso L, Girolamo F, Rizzi A, Virgintino D,Roncali L, Ambrosi G (2004) Glutamic acid decarboxylaseimmunoreactive large neuron types in the granular layer of the hu-man cerebellar cortex. Anat Embryol (Berlin) 208:55–64

Floris A, Dino M, Jacobowitz DM, Mugnaini E (1994) The unipolarbrush cells of the rat cerebellar cortex and cochlear nucleus arecalretinin-positive: a study by light and electron microscopicimmunocytochemistry. Anat Embryol (Berlin) 189:495–520

Fogarty M, Grist M, Gelman D, Marin O, Pachnis V, Kessaris N(2007) Spatial genetic patterning of the embryonic neuroepithe-lium generates GABAergic interneuron diversity in the adult cor-tex. J Neurosci 27:10935–10946

Fox CA, Hillman DE, Siegesmund KA, Dutta CR (1967) The primatecerebellar cortex: a Golgi and electron microscopic study. ProgBrain Res 25:174–225

Fremeau RT Jr, Kam K, Qureshi T, Johnson J, Copenhagen DR,Storm-Mathisen J, Chaudhry FA, Nicoll RA, Edwards RH (2004)Vesicular glutamate transporters 1 and 2 target to functionallydistinct synaptic release sites. Science 304:1815–1819

Fujita S (1967) Quantitative analysis of cell proliferation and diVeren-tiation in the cortex of the postnatal mouse cerebellum. J Cell Biol32:277–287

Fukumitsu H, Ohtsuka M, Murai R, Nakamura H, Itoh K, Furukawa S(2006) Brain-derived neurotrophic factor participates in determi-nation of neuronal laminar fate in the developing mouse cerebralcortex. J Neurosci 26:13218–13230

Geurts FJ, De Schutter E, Dieudonne S (2003) Unraveling the cerebel-lar cortex: cytology and cellular physiology of large-sized inter-neurons in the granular layer. Cerebellum 2:290–299

Geurts FJ, Timmermans J, Shigemoto R, De Schutter E (2001) Mor-phological and neurochemical diVerentiation of large granularlayer interneurons in the adult rat cerebellum. Neuroscience104:499–512

Gilbertson RJ, Ellison DW (2008) The origins of medulloblastomasubtypes. Annu Rev Pathol 3:341–365

Glickstein SB, Moore H, Slowinska B, Racchumi J, Suh M, ChuhmaN, Ross ME (2007) Selective cortical interneuron and GABAdeWcits in cyclin D2-null mice. Development 134:4083–4093

Gliem M, Weisheit G, Mertz KD, Endl E, Oberdick J, Schilling K (2006)Expression of classical cadherins in the cerebellar anlage: Quantita-tive and functional aspects. Mol Cell Neurosci 33:447–458

Goldowitz D, Hamre K (1998) The cells and molecules that make acerebellum. TINS 21:375–382

Golgi C (1903) Sulla Wna anatomia del cervelletto umano. Lecture, Isti-tuto Lombardo di Sci. e Lett. 8 January 1874. In: Opera Omnia, Ist-ologia normale, chap. V, vol I. Ulrico Hoepli, Milano, pp 99–111

Hallonet MER, Le Douarin NM (1993) Tracing neuroepithelial cells ofthe mesencephalic and metencephalic alar plates during cerebellarontogeny in quil-chick chimaeras. Eur J Neurosci 5:1145–1155

Harris J, Moreno S, Shaw G, Mugnaini E (1993) Unusual neuroWla-ment composition in cerebellar unipolar brush neurons. J Neuro-cytol 22:1039–1059

Hecker D, Kappler J, Glassmann A, Schilling K, Alt W (2008) Imageanalysis of time-lapse movies-A precision control guided approachto correct motion artefacts. J Neurosci Methods 172:67–73

Hoshino M, Nakamura S, Mori K, Kawauchi T, Terao M, NishimuraYV, Fukuda A, Fuse T, Matsuo N, Sone M, Watanabe M, Bito H,Terashima T, Wright CV, Kawaguchi Y, Nakao K, Nabeshima Y(2005) Ptf1a, a bHLH transcriptional gene, deWnes GABAergicneuronal fates in cerebellum. Neuron 47:201–213

Huard JMT, Forster CC, Carter ML, Sicinski P, Ross ME (1999) Cer-ebellar histogenesis is disturbed in mice lacking cyclin D2. Devel-opment 126:1927–1935

Ilijic E, Guidotti A, Mugnaini E (2005) Moving up or moving down?Malpositioned cerebellar unipolar brush cells in reeler mouse.Neuroscience 136:633–647

Ito M (2008) Control of mental activities by internal models in the cer-ebellum. Nat Rev Neurosci 9:304–313

Jaarsma D, Wenthold RJ, Mugnaini E (1995) Glutamate receptor sub-units at mossy Wber-unipolar brush cell synapses: light and elec-tron microscopic immunocytochemical study in cerebellar cortexof rat and cat. J Comp Neurol 357:145–160

Koscheck T, Weyer A, Schilling RL, Schilling K (2003) Morphologi-cal development and neurochemical diVerentiation of cerebellarinhibitory interneurons in microexplant cultures. Neuroscience116:973–984

Kozmik Z, Sure U, Ruedi D, Busslinger M, Aguzzi A (1995) Deregu-lated expression of PAX5 in medulloblastoma. Proc Natl AcadSci USA 92:5709–5713

Kunzelmann P, Blumcke I, Traub O, Dermietzel R, Willecke K (1997)Coexpression of connexin45 and -32 in oligodendrocytes of ratbrain. J Neurocytol 26:17–22

Laine J, Axelrad H (1994) The candelabrum cell: a new interneuron inthe cerebellar cortex. J Comp Neurol 339:159–173

Laine J, Axelrad H (1996) Morphology of the Golgi-impregnated Lug-aro cell in the rat cerebellar cortex: a reappraisal with a descrip-tion of its axon. J Comp Neurol 375:618–640

Laine J, Axelrad H (1998) Lugaro cells target basket and stellate cellsin the cereballer cortex. NeuroReport 13:2399–2403

Laine J, Axelrad H (2002) Extending the cerebellar Lugaro cell class.Neuroscience 115:363–374

Landau E (1927) Beitrag zur Kenntnis der Körnerschicht des Kle-inhirns. Anat Anz 62:391–398

Lander C, Kind P, Maleski M, HockWeld S (1997) A family of activity-dependent neuronal cell-surface chondroitin sulfate proteogly-cans in cat visual cortex. J Neurosci 17:1928–1939

Lander C, Zhang H, HockWeld S (1998) Neurons produce a neuronalcell surface-associated chondroitin sulfate proteoglycan. J Neuro-sci 18:174–183

Lein ES, Hawrylycz MJ, Ao N, Ayres M, Bensinger A, Bernard A et al(2007) Genome-wide atlas of gene expression in the adult mousebrain. Nature 445:168–176

Leone DP, Srinivasan K, Chen B, Alcamo E, McConnell SK (2008)The determination of projection neuron identity in the developingcerebral cortex. Curr Opin Neurobiol 18:28–35

Leto K, Bartolini A, Rossi F (2008) Development of cerebellar GAB-Aergic interneurons: origin and shaping of the minibrain cocalconnections. The Cerebellum (in press)

Leto K, Carletti B, Williams IM, Magrassi L, Rossi F (2006) DiVerenttypes of cerebellar GABAergic interneurons originate from acommon pool of multipotent progenitor cells. J Neurosci26:11682–11694

Li J, Lao Z, Joyner A (2002) Changing requirements for gbx2 in devel-opment of the cerebellum and maintenance of the mid/hindbrainorganizer. Neuron 36:31–43

Li S, Tokuyama T, Yamamoto J, Koide M, Yokota N, Namba H (2005)Potent bystander eVect in suicide gene therapy using neural stemcells transduced with herpes simplex virus thymidine kinase gene.Oncology 69:503–508

Lind D, Franken S, Kappler J, Jankowski J, Schilling K (2004) Char-acterization of the neuronal marker NeuN as a multiply phosphor-ylated antigen with discrete subcellular localization. J NeurosciRes 79:295–302

Lugaro E (1894) Sulle connessioni tra gli elementi nervosi della cor-teccia cerebellare con considerazioni generali sul signiWcato Wsi-ologico dei rapporti tra gli elementi nervosi. Riv Sper Fren MedLeg 20:297–331

Magdaleno S, Jensen P, Brumwell CL, Seal A, Lehman K, Asbury A,Cheung T, Cornelius T, Batten DM, Eden C, Norland SM, RiceDS, Dosooye N, Shakya S, Mehta P, Curran T (2006) BGEM: AnIn Situ Hybridization Database of Gene Expression in the Embry-onic and Adult Mouse Nervous System. PLoS Biol 4:e86

123


Maricich SM, Herrup K (1999) Pax-2 expression deWnes a subset ofGABAergic interneurons and their precursors in the developingmurine cerebellum. J Neurobiol 41:281–294

Martinez-Maza R, Poyatos I, Lopez-Corcuera B, Unez N, Gimenez C,Zafra F, Aragon C (2001) The role of N-glycosylation in transportto the plasma membrane and sorting of the neuronal glycine trans-porter GLYT2. J Biol Chem 276:2168–2173

Mathis C, Collin L, Borrelli E (2003) Oligodendrocyte ablation im-pairs cerebellum development. Development 130:4709–4718

Mathis L, Bonnerot C, Puelles L, Nicolas JF (1997) Retrospectiveclonal analysis of the cerebellum using genetic laacZ/lacZ mousemosaics. Development 124:4089–4104

Matthews RT, Kelly GM, Zerillo CA, Gray G, Tiemeyer M, HockWeldS (2002) Aggrecan glycoforms contribute to the molecular heter-ogeneity of perineuronal nets. J Neurosci 22:7536–7547

Maxeiner S, Kruger O, Schilling K, Traub O, Urschel S, Willecke K(2003) Spatiotemporal transcription of connexin45 during braindevelopment results in neuronal expression in adult mice. Neuro-science 119:689–700

McConnell SK, Kaznowski CE (1991) Cell cycle dependence of lam-inar determination in developing neocortex. Science 254:282–285

Mertz K, Koscheck T, Schilling K (2000) Brain derived neuronotroph-ic factor modulates dendritic morphology of cerebellar basket andstellate cells. Neuroscience 97:303–310

Mesnil M, Yamasaki H (2000) Bystander eVect in herpes simplex vi-rus-thymidine kinase/ganciclovir cancer gene therapy: role ofgap-junctional intercellular communication. Cancer Res60:3989–3999

Meyyappan M, Wong H, Hull C, Riabowol KT (1998) Increasedexpression of cyclin D2 during multiple states of growth arrest inprimary and established cells. Mol Cell Biol 18:3163–3172

Miale IR, Sidman RL (1961) An autoradiographic analysis of histo-genesis in the mouse cerebellum. Exp Neurol 4:277–296

Millen KJ, Gleeson JG (2008) Cerebellar development and disease.Curr Opin Neurobiol 18:12–19

Milosevic A, Goldman JE (2002) Progenitors in the postnatal cerebel-lar white matter are antigenically heterogeneous. J Comp Neurol452:192–203

Milosevic A, Goldman JE (2004) Potential of progenitors from postna-tal cerebellar neuroepithelium and white matter: lineage speciWedvs.multipotent fate. Mol Cell Neurosci 26:342–353

Morin F, Dino MR, Mugnaini E (2001) Postnatal diVerentiation of uni-polar brush cells and mossy Wber-unipolar brush cell synapses inrat cerebellum. Neuroscience 104:1127–1139

Mugnaini E (1972) The histology and cytology of the cerebellar cortex.In: Larsell O, Jansen J (eds) The comparative histology of the cer-ebellum. The humal cerbellum, cerebellar connections and cere-bellar cortex. University of Minnesota Press, Minneapolis, pp201–264

Mugnaini E, Dino MR, Jaarsma D (1997) The unipolar brush cells ofthe mammalian cerebellum and cochlear nucleus: cytology andmicrocircuitry. Prog Brain Res 114:131–150

Mugnaini E, Floris A (1994) The unipolar brush cell: a neglected neuronof the mammalian cerebellar cortex. J Comp Neurol 339:174–180

Mugnaini E, Floris A, Wright-Goss M (1994) Extraordinary synapsesof the unipolar brush cell: An electron microscopic study in the ratcerebellum. Synapse 16:284–311

Muth TR, Ahn J, Caplan MJ (1998) IdentiWcation of sorting determi-nants in the C-terminal cytoplasmic tails of the gamma-aminobu-tyric acid transporters GAT-2 and GAT-3. J Biol Chem273:25616–25627

Nakhai H, Sel S, Favor J, Mendoza-Torres L, Paulsen F, Duncker GI,Schmid RM (2007) Ptf1a is essential for the diVerentiation ofGABAergic and glycinergic amacrine cells and horizontal cells inthe mouse retina. Development 134:1151–1160

Napieralski JA, Eisenman LM (1993) Developmental analysis of theexternal granule cell layer in the meander tail mutant mouse: docerebellar microneurons have independent progenitors? Dev Dyn197:244–254

Nunzi MG, Birnstiel S, Bhattacharyya BJ, Slater NT, Mugnaini E(2001) Unipolar brush cells form a glutamatergic projection sys-tem within the mouse cerebellar cortex. J Comp Neurol 434:329–341

Nunzi MG, Russo M, Mugnaini E (2003) Vesicular glutamate trans-porters VGLUT1 and VGLUT2 deWne two subsets of unipolarbrush cells in organotypic cultures of mouse vestibulocerebellum.Neuroscience 122:359–371

Nunzi MG, Shigemoto R, Mugnaini E (2002) DiVerential expressionof calretinin and metabotropic glutamate receptor mGluR1alphadeWnes subsets of unipolar brush cells in mouse cerebellum. JComp Neurol 451:189–199

Oberdick J, Baader SL, Schilling K (1998) From zebra stripes to postalzones: deciphering patterns of gene expression in the cerebellum.TINS 21:383–391

Ottersen OP, Davanger S, Storm-Mathisen J (1987) Glycine-likeimmunoreactivity in the cerebellum of rat and Senegalese ba-boon, Papio papio: a comparison with the distribution of GABA-like immunoreactivity and with [3H]glycine and [3H]GABA up-take. Exp Brain Res 66:211–221

Palay SL, Chan-Palay V (1974) Cerebellar cortex. Cytology and orga-nization. Springer-Verlag, Berlin

Pascual M, Abasolo I, Mingorance-Le Meur A, Martinez A, Del RioJA, Wright CV, Real FX, Soriano E (2007) Cerebellar GABAer-gic progenitors adopt an external granule cell-like phenotype inthe absence of Ptf1a transcription factor expression. Proc NatlAcad Sci USA 104:5193–5198

Pietsch T, Taylor MD, Rutka JT (2004) Molecular pathogenesis ofchildhood brain tumors. J Neurooncol 70:203–215

Porrill J, Dean P, Stone JV (2004) Recurrent cerebellar architecturesolves the motor-error problem. Proc Biol Sci 271:789–796

Ramon y Cajal S (1899) Textura del Sistema Nervioso del Hombre yde los Vertebrados. Moya, Madrid

Ramon y Cajal S (1909) Histologie du système nerveux de l’homme etdes vertébrés. II. A. Maloine, Paris

Rong Y, Wang T, Morgan JI (2004) IdentiWcation of candidate Pur-kinje cell-speciWc markers by gene expression proWling in wild-type and pcd(3 J) mice. Mol Brain Res 132:128–145

Rossi DJ, Alford S, Mugnaini E, Slater NT (1995) Properties of trans-mission at a giant glutamatergic synapse in cerebellum: the mossyWger-unipolar brush cell synapse. J Neurophysiol 74:24–42

Sahin M, HockWeld S (1990) Molecular identiWcation of the Lugarocell in the cat cerebellar cortex. J Comp Neurol 301:575–584

Saitow F, Murakoshi T, Suzuki H, Konishi S (2005) Metabotropic P2Ypurinoceptor-mediated presynaptic and postsynaptic enhance-ment of cerebellar GABAergic transmission. J Neurosci 25:2108–2116

Schuske K, Jorgensen EM (2004) Vesicular glutamate transporter–shooting blanks. Science 304:1750–1752

Sekerkova G, Ilijic E, Mugnaini E (2004) Time of origin of unipolarbrush cells in the rat cerebellum as observed by prenatal bromo-deoxyuridine labeling. Neuroscience 127:845–858

Shimono T, Nosaka S, Sasaki K (1976) Electrophysiological study onthe postnatal development of neuronal mechanisms in the rat cer-ebellar cortex. Brain Res 108:279–294

Simat M, Parpan F, Fritschy JM (2007) Heterogeneity of glycinergicand gabaergic interneurons in the granule cell layer of mouse cer-ebellum. J Comp Neurol 500:71–83

Sotelo C (1990) Cerebellar synaptogenesis: what we can learn frommutant mice. J Exp Biol 153:225–249

Sotelo C (2004) Cellular and genetic regulation of the development ofthe cerebellar system. Prog Neurobiol 72:295–339

123


Sultan F, Bower JM (1998) Quantitative Golgi study of the rat cerebel-lar molecular layer interneurons using principal component anal-ysis. J Comp Neurol 393:353–373

Takacs J, Borostyankoi ZA, Veisenberger E, Vastagh C, Vig J,Gorcs TJ, Hamori J (2000) Postnatal development of unipolarbrush cells in the cerebellar cortex of cat. J Neurosci Res61:107–115

Tanaka I, Ezure K (2004) Overall distribution of GLYT2 mRNA-con-taining versus GAD67 mRNA-containing neurons and colocal-ization of both mRNAs in midbrain, pons, and cerebellum in rats.Neurosci Res 49:165–178

Tidcombe H, Jackson-Fisher A, Mathers K, Stern DF, Gassmann M,Golding JP (2003) Neural and mammary gland defects in ErbB4knockout mice genetically rescued from embryonic lethality. ProcNatl Acad Sci USA 100:8281–8286

Triller A, Cluzeaud F, Korn H (1987) Gamma-aminobutyric acid-con-taining terminals can be apposed to glycine receptors at centralsynapses. J Cell Biol 104:947–956

Uhl M, Weiler M, Wick W, Jacobs AH, Weller M, Herrlinger U (2005)Migratory neural stem cells for improved thymidine kinase-basedgene therapy of malignant gliomas. Biochem Biophys Res Com-mun 328:125–129

Vig J, Goldowitz D, Steindler DA, Eisenman LM (2005a) Compart-mentation of the reeler cerebellum: Segregation and overlap ofspinocerebellar and secondary vestibulocerebellar Wbers and theirtarget cells. Neuroscience 130:735–744

Vig J, Takacs J, Abraham H, Kovacs GG, Hamori J (2005b) Calretinin-immunoreactive unipolar brush cells in the developing humancerebellum. Int J Dev Neurosci 23:723–729

Vilz TO, Moepps B, Engele J, Molly S, Littman DR, Schilling K(2005) The SDF-1/CXCR4 pathway and the development of thecerebellar system. Eur J NeuroSci 22:1831–1839

Visel A, Thaller C, Eichele G (2004) GenePaint.org: an atlas of geneexpression patterns in the mouse embryo. Nucleic Acids Res32:D552–D556

Voogd J, Glickstein M (1998) The anatomy of the cerebellum. TINS21:370–375

Weisheit G, Gliem M, Endl E, PfeVer PL, Busslinger M, Schilling K(2006) Postnatal development of the murine cerebellar cortex:formation and early dispersal of basket, stellate and Golgi neu-rons. Eur J NeuroSci 24:466–478

Weyer A, Schilling K (2003) Developmental and cell type-speciWcexpression of the neuronal marker NeuN in the murine cerebel-lum. J Neurosci Res 73:400–409

Wojcik SM, Katsurabayashi S, Guillemin I, Friauf E, Rosenmund C,Brose N, Rhee JS (2006) A shared vesicular carrier allows synap-tic corelease of GABA and glycine. Neuron 50:575–587

Xiang Z, Burnstock G (2005) Changes in expression of P2X purino-ceptors in rat cerebellum during postnatal development. DevBrain Res 156:147–157

Yamanaka H, Yanagawa Y, Obata K (2004) Development of stellateand basket cells and their apoptosis in mouse cerebellar cortex.Neurosci Res 50:13–22

Zervas M, Millet S, Ahn S, Joyner AL (2004) Cell behaviors and ge-netic lineages of the mesencephalon and rhombomere 1. Neuron43:345–357

Zhang L, Goldman JE (1996a) Developmental fates and migratorypathways of dividing progenitors in the postnatal rat cerebellum.J Comp Neurol 370:536–550

Zhang L, Goldman JE (1996b) Generation of cerebellar interneuronsfrom dividing progenitors in white matter. Neuron 16:47–54

Zordan P, Croci L, Hawkes R, Consalez GG (2008) Comparative anal-ysis of proneural gene expression in the embryonic cerebellum.Dev Dyn 237:1726–1735

123

besides purkinje cells and granule neurons: an appraisal ... · located in the granule cell layer...

Documents